fixes

artemis/cogs/chat.py

@@ -1,5 +1,6 @@
 from __future__ import annotations
 
+import asyncio
 import logging
 import re
 from typing import TYPE_CHECKING
@@ -42,7 +43,6 @@ emoji_map = {
     "😛": ":P",
     "🤑": "$-$",
     "🤗": "(hug)",
-    "🤔": ":/",
     "😎": "8)",
     "😏": "^_^",
     "😒": "-_-",
@@ -99,9 +99,10 @@ ARTEMIS_RE = re.compile(r"\bar(i)?temis\b", flags=re.IGNORECASE)
 class Chat(commands.Cog):
     def __init__(self, bot: Artemis):
         self.bot: Artemis = bot
-        self.memory: dict[int, list[dict]] = {}
+        self.memory: list[dict] = []
         self.model = "google/gemma-2-2b-it"
         self.client = AsyncInferenceClient(api_key=self.bot.secrets.huggingface)
+        self.lock = asyncio.Lock()
 
     def replace_emojis(self, text: str) -> str:
         return EMOJI_RE.sub(lambda match: emoji_map[match.group(0)], text)
@@ -114,40 +115,44 @@ class Chat(commands.Cog):
             content = content.replace(user.mention, "@" + user.display_name)
         return content
 
-    def get_memory(self, user_id: int):
-        if user_id not in self.memory:
-            self.memory[user_id] = []
-        return self.memory[user_id]
+    def add_memory(self, role: str, message: str):
+        prompt = "You're Artemis, a friendly AI hanging out in this Discord server, following is a user chat message directed at you.\n\n"
+        if len(self.memory) == 0:
+            message = prompt + message
+        if len(self.memory) >= 15:
+            del self.memory[0]
+            del self.memory[0]
+            self.memory[0] = {"role": "user", "content": prompt + self.memory[0]["content"]}
+        self.memory.append({"role": role, "content": message})
 
-    def add_memory(self, user_id: int, role: str, message: str):
-        memory = self.get_memory(user_id)
-        if len(memory) == 0:
-            message = f"You're Artemis, a friendly AI hanging out in this Discord server, following is a user chat message directed at you.\n\n{message}"
-        if len(memory) >= 10:
-            del memory[1]
-        memory.append({"role": role, "content": message})
-        self.memory[user_id] = memory
+    def add_user_memory(self, message: str):
+        self.add_memory("user", message)
 
-    def add_user_memory(self, user_id: int, message: str):
-        self.add_memory(user_id, "user", message)
+    def add_assistant_memory(self, message: str):
+        self.add_memory("assistant", message)
 
-    def add_assistant_memory(self, user_id: int, message: str):
-        self.add_memory(user_id, "assistant", message)
-
-    async def chat(self, user_id: int, message: str):
-        self.add_user_memory(user_id, message)
+    async def chat(self, message: str):
+        self.add_user_memory(message)
         response = await self.client.chat.completions.create(
             model=self.model,
-            messages=self.get_memory(user_id),
+            messages=self.memory,
             max_tokens=500,
             stream=False,
         )
         chat_response = response.choices[0].message.content
-        self.add_assistant_memory(user_id, chat_response)
+
+        chat_response = self.replace_emojis(chat_response)
+        chat_response = self.strip_emojis(chat_response)
+        chat_response = re.sub(r"[ ]{2,}", " ", chat_response)
+
+        self.add_assistant_memory(chat_response)
         return chat_response
 
     @commands.Cog.listener()
     async def on_message(self, message: discord.Message):
+        if message.author.bot:
+            return
+
         reference = message.reference
         parent = reference.cached_message if reference else None
 
@@ -173,10 +178,8 @@ class Chat(commands.Cog):
 
         try:
             async with message.channel.typing():
-                response = await self.chat(message.author.id, content)
-                response = self.replace_emojis(response)
-                response = self.strip_emojis(response)
-                response = re.sub(r"[ ]{2,}", " ", response)
+                async with self.lock:
+                    response = await self.chat(content)
             await message.reply(response)
         except aiohttp.client_exceptions.ClientResponseError as err:
             await message.reply(f"{self.model} error: {err.status} {err.message}")